Pragmatic Policies to Tackle VAT Fraud in the European Union

Articles Fabrizio Borselli* Pragmatic Policies to Tackle VAT Fraud in the European Union In recent decades, the Member States of the European Union h...
3 downloads 1 Views 248KB Size
Articles Fabrizio Borselli*

Pragmatic Policies to Tackle VAT Fraud in the European Union In recent decades, the Member States of the European Union have registered a huge increase of spectacular and highly organized VAT fraud. Whilst the tax authorities are fighting hard to stop this phenomenon, fraud has become an established industry. In this article, the author analyses several measures to tackle VAT fraud. 1. Introduction Indirect taxes are a major source of revenue in most countries and, in the European Union, also an important policy instrument for achieving a greater integration of the economies of the individual Member States. The European VAT system has been harmonized over time, at least as regards its base of assessment. Its typical feature is the mechanism of deduction of input tax,1 which, on the one hand, ensures the economic efficiency of the tax system but, on the other hand, is the source of sophisticated tax fraud.2 In recent decades, the Member States have registered a huge increase of spectacular and highly organized VAT fraud. Whilst the tax authorities are fighting hard to stop this phenomenon, fraud has become an established industry,3 facilitated by the inadequacy of the national control systems, lack of effective regulatory mechanisms, and delay in the transition of the VAT system to an origin-based type of tax in respect of intraCommunity transactions. Tax fraud is an obstacle to the smooth functioning of the internal market. It prevents fair competition, erodes the revenues of the Member States and negatively affects the Community’s own resources. Since the governments of the Member States are forced to make up the shortfall in revenue due to fraud, an increased level of fraud results in a heavier tax burden and imposition of more stringent obligations on legitimate businesses. The fight against tax fraud in the European Union is part of the Lisbon strategy and is supported by the VAT Directive. The functioning of the internal market is dependent on the efforts of the individual Member States to develop more sophisticated control and intelligence systems, and of the international bodies in charge of preventing money laundering. The purpose of this article is twofold. Firstly, it outlines the fraudulent VAT schemes that pose the most serious threat to the functioning of the VAT system, and evaluates the effectiveness of the strategies adopted to combat and prevent that type of fraud. Secondly, it comments on the costs and benefits of more far-reaching measures to be implemented in the near future at EU and national © IBFD

level. The purpose of this article is not to analyse the mechanics of “basic” fraudulent schemes (for example the exchange of fake invoices, abuse of the right to deduct input tax and fraudulent VAT returns). 2. Evolution of Fraud Schemes and Related Impact on Tax Revenue In VAT fraud schemes, a taxable person (“missing trader”) acquires goods from another Member State free of VAT, which is in accordance with the destination principle. Subsequently, the missing trader resells the goods at a VAT-inclusive price, but does not remit to the national tax authorities the VAT collected from his customer. This simple scheme (“acquisition fraud”), which enables the supplier to supply the goods at a low price by embezzling the VAT payable to the tax authorities, has gradually assumed more sophisticated forms. The well-known “carousel fraud” implies that the same goods are repeatedly supplied as a cross-border transaction, in a “circular pattern” (between the same parties), enabling the fraudsters to steal VAT from the authorities at each “turn”.4 It is not a form of tax avoidance or advanced tax planning but plainly large-scale theft and, as such, a deliberate and systematic attack on the tax system and government revenues, in the framework of which the criminals may involve a large number of other parties, and commit various offences, such as filing fraudulent VAT returns and criminal association. Because of the closed and contrived nature of the transactions involved in the fraud, the amounts of money that can be stolen are in theory unlimited.5 In order to cover up the scheme, the fraudsters may set up extensive and complicated chains of transactions involving parties in several Member States. In principle, VAT carousels may involve any type of goods, although what the goods often have in common *

Official at Bank of Italy, Tax Department. The article is based on the author’s study on tax compliance and VAT fraud. The views expressed in this article represent only those of the author and place no responsibility on the Bank of Italy. 1. M. Keen and B. Lockwood, “Is the VAT a Money Machine?”, National Tax Journal, Vol. 59 (2006), pp. 905-928. 2. M. Keen and S. Smith, “VAT Fraud and Evasion: What Do We Know, and What Can Be Done?”, IMF Working Papers (2007), WP/07/31. 3. International VAT Association, “Combating VAT fraud in the European Union. The way forward”, Report to the European Commission, March 2007. 4. The fraud is similar to the acquisition fraud in the early stages, but the goods are sold through a series of companies in a certain country and then reexported to another EU Member State, hence the goods move in a “carousel”. The exporter, on his part, claims back the input VAT paid to his supplier and, as a result, the tax authorities suffer a loss. 5. Financial Action Task Force, “Laundering the proceeds of VAT carousel fraud”, 23 February 2007, pp. 2, 18, 22. INTERNATIONAL VAT MONITOR SEPTEMBER/OCTOBER 2008

333

Articles

is that they are readily available in large quantities, have a high value and low weight, and become rapidly technically obsolete.6 Fraud can also be committed through completely fictitious transactions involving innocent businesses or by contrast, businesses that are fully aware of the irregular nature of the transaction. Fraudsters commonly target their activities at industries characterized by a significant grey market. The “missing traders” are empty boxes formally controlled by dummy managers. These short-lived companies (which may last for a few weeks or, at the most, a few months) carry out simulated activities, managed by persons who move, if necessary, their place of business abroad and transfer the proceeds of their criminal activities to tax havens. In this context, “buffers”,7 carriers, VAT warehouses and legal firms play an important role. Services increasingly form part of VAT fraud schemes, as they are capable of covering up illegal supplies of goods. Indeed, a single business may supply both goods and services, to the effect that VAT claims resulting from the fraudulent activities relating to goods are compensated by VAT liabilities resulting from the supply of services. The complexity of the schemes and the involvement of third parties make the fraud even more opaque and the measures taken by the tax authorities less effective. Whilst, originally, most carousel chains were relatively simple and only involved EU Member States, an increasing number of recently discovered fraud cases involve third countries, such as Dubai, Hong Kong, Switzerland, the United States and the Arab Emirates.8 Under the Community customs transit rules,9 goods may be imported into the European Union and, subsequently, move within the territory of the European Union free of customs duty and VAT until they reach the Member State of final destination. After the fraud has taken place, the goods are re-exported to the third country of origin and the carousel starts again, leaving the import duties and import VAT unpaid. Alternatively, the imported goods, before reaching their supposed final destination, are diverted to the black market (acquisition fraud). A different current scheme involves non-EU countries with low labour costs and raw materials costs. Goods originating from those countries are imported into the European Union at a declared customs value, which is considerably below the actual value of the imported goods, often even below the world market price of the materials incorporated in the goods. This undervaluation, which is based on fake invoices, causes a sharp reduction of the taxes and duties due on the importation of the goods and enables the importers to commit carousel or acquisition fraud. The subsequent domestic supplies of the goods are normally fictitious and the proceeds of the fraud are channelled back to the country of origin of the goods and are used to finance other criminal operations. Thanks to the huge volume of international trade, criminals take root in port areas, purchase terminals, shipping companies and logistic centres, and expand their activities to chains of stores, undermining the competitiveness of the market and harming national 334

INTERNATIONAL VAT MONITOR SEPTEMBER/OCTOBER 2008

production.10 This phenomenon is likely to put pressure on the customs authorities to further ease their supervision of goods under customs control. Profits generated by VAT fraud leave the European Union and, disguised as the proceeds derived from transactions between legitimate businesses, are laundered through off-shore banking circuits. They are then used to pay the persons involved in the fraud and those who have facilitated it. The schemes carefully avoid the main actors (missing traders) formally becoming the owners of assets (real estate, cash, etc.), which, in the worst-case scenario, could be seized by the tax authorities, or reveal their true identity to the banks involved. Fraudsters frequently use the same bank account and transfer the money online, maintaining their anonymity to the largest possible extent. Laundering substantial sums of money derived from VAT fraud requires expert knowledge of financial procedures, which can only be found with sophisticated organized criminal gangs, who are attracted by the large profits available at a relatively low risk. It has been established that there are links between the laundering of the proceeds derived from VAT fraud and the funds needed to finance other forms of organized crime; the proceeds derived from carousel fraud may have become the most important source of financing other criminal activities.11 Setting an exact figure of the size of VAT fraud is extremely difficult.12 Several national studies on fraud give estimates that are, at best, rough approximations of the size of the phenomenon, which do not allow one to draw clear conclusions.13 Furthermore, carousel fraud itself impedes accurate measurement of trade flows.14

6. For example, computers and mobile phones, but also meat, cars, mineral oils, soft drinks, cosmetics, precious metals and, more recently, services. 7. A “buffer company” is generally a fully compliant trader carrying out regular business outside the fraud, but also being drawn into fraudulent activity by generating additional transactions. 8. House of Lords, “Stopping the Carousel: Missing Trader Fraud in the EU”, European Union Committee, 20th Report of Session 2006-07, HL Paper 101 (2007), 25 May 2007. 9. Community transit rules facilitate the movement of goods throughout the territory of the European Union by temporarily suspending the duties and other charges on goods imported into the Community, until they reach their final destination. Customs formalities are shifted from the border to the place of final destination. 10. Italian Customs Agency, Oltrefrontiera, Bimestrale d’informazione, year IV, No. 4, December 2006/March 2007; Circular 45/D of 17 November 2005. 11. Financial Action Task Force, see note 5, p. 18. 12. VAT fraud, together with tax avoidance and other non-compliant activities, contributes to the overall “VAT gap”, which is the difference between the net theoretical tax liabilities and actual VAT revenue. The VAT gap can be estimated using macroeconomics (“top-down” approach) or microeconomics data (“bottom-up”). 13. OECD, “VAT fraud and the trade supply chain: the “missing trader” or carousel fraud and its impact on trade statistics”, National Accounts and Economic Statistics – International Trade Statistics, Statistics Directorate, 7 September 2006. 14. Bank of England, “Inflation Report” (2006), August 2006, www.bankofengland.co.uk. While the raw data on imports from the European Union excludes activity in fraudulent goods, since the missing traders do not submit any VAT declarations, the raw data on exports normally includes the fraudulent transactions. See also D. Ruffless, G. Tily, D. Caplan and S. Tudor, “VAT missing trader intra-Community fraud: the effect on Balance of Payments statistics and UK National Accounts”, Economic Trends, No. 597 (2003), August 2003.

© IBFD

Articles Figure 1

Carousel fraud: VAT revenue losses (2006) 14,8

6,0

5,05 5,0 3,8 4,0

3,0

2,6

2,3 2,0

2,0

1,5

EU carousel fraud

Others

0,2 Luxembourg

Portugal

0,2 Belgium

0,4

“Third-countries" carousel fraud

Finally, the size of the Member State, as well as its VAT repayment policy and its control measures, has a strong influence on the level of fraud. Also, fraudulent activities can vary significantly from year to year. According to published data, the loss of VAT in the European Union would range between EUR 60 and 100 billion per year.15 In some Member States, the loss of revenue due to carousel fraud would reach 10% of the total VAT revenue.16 In Germany, the loss of revenue due to VAT fraud is estimated at between EUR 15 and 18 billion,17 or about 11% of the total VAT revenue,18 of which EUR 4.5 billion is attributed to carousel fraud.19 In the United Kingdom, attempted intra-Community VAT fraud involving a missing trader would range between GBP 2.25 and 3.25 billion, with an actual impact on VAT revenue of between GBP 1 and 2 billion.20 Another source estimates the EU annual VAT lost due to carousel fraud (excluding fraud involving third-countries),21 at EUR 14.8 billion, of which about EUR 3.8 billion (or EUR 8.85 billion, including the loss due to fraud involving third countries) is attributed to the United Kingdom alone; see Figure 1. 3. Effectiveness of the EU and National Strategies, and VAT Fraud Trends The abolition of the internal tax borders was accompanied by the introduction of new instruments aimed at enhancing administrative cooperation between the Member States by means of a system for the automatic exchange of information (VIES) and new Community rules on mutual assistance. © IBFD

0,7 0,4

Denmark

France

Germany

Italy

Spain

United Kingdom

EU

0,0

Netherlands

0,7

1,0

The enactment in 2004 of a new EU Regulation on administrative cooperation22 speeded up and enhanced cross-border assistance between Member States and increased the efficiency of controls by means of clearer procedures and more direct links between the local tax offices in different Member States. Other initiatives have been taken in the form of a spontaneous exchange of information between the intelligence units of the Member States (for example, implementation of the network Eurocanet – European Carousel Network – sponsored

15. International VAT Association, see note 3; Kovacs, “Introductory remarks”, Press conference on the adoption of the Communication on Fraud, Brussels, 31 May 2006. 16. Kovacs, “Open speech”, Conference “Tackling VAT fraud: possible ways forward”, Brussels, 29 February 2007. 17. Monatsbericht des BMF, January 2006, p. 45, mentioned in Court of Auditors, “Special report No. 11/2006 on the Community transit system, together with the Commission’s replies”, (2007/C 44/01), OJ C 44/1 of 27 February 2007; International VAT Association, see note 3; Financial Times.com, Germany blocks EU deal on e-commerce services, 8 June 2006. 18. See Tax Notes International, Tax Analysts, August 2005, p. 489. 19. European Commission, “Tackling VAT fraud: possible ways forward”, Document for the workshop, TAXUD, 2007. 20. HM Revenue & Customs, “Measuring indirect tax losses – 2007”, October 2007. 21. Extracts from “Attempt to measure cross border VAT fraud. Breakdown within the EU – OCS SPF Finances Belgique”, in House of Lords (see note 8), pp. 126-132. The overall amount is an arithmetic average of the results obtained by four different macroeconomic approaches and should be treated as an “upper limit” valuation. 22. This is the exchange of information between Member States whereby tax authorities assist each other and cooperate with the Commission, pursuant to Regulation (EC) No. 1798/2003, in order to ensure the proper application of VAT. The cooperation takes place both on request (through the VAT Information Exchange System, VIES) and without prior request (mutual assistance between Member States).

INTERNATIONAL VAT MONITOR SEPTEMBER/OCTOBER 2008

335

Articles

by Belgium and supported by the Commission and OLAF23).

principles of “tax neutrality”29 and “proportionality” have particular relevance.

As far as customs is concerned, the European Commission has made a considerable effort to introduce, with effect from 2006, a computerized system (the NCTS) to monitor the transit of goods under customs control, which replaced the previous paper-based procedure and provides for an electronic exchange of messages between the customs offices and between traders and customs offices. This system is designed to prevent traditional forms of fraud based on forged documents and allows real-time identification of missing goods, enabling an immediate start of the assessment procedure.

The use of the reverse charge mechanism in respect of business-to-business (B2B) transactions – shifting the VAT liability to the customer, rather than leaving it with the supplier – may represent a more radical strategy, as the missing trader would have no VAT liability at all. The mechanism generally applies in sectors characterized by a considerable risk of fraud. Nevertheless, the reverse charge mechanism may increase the traders’ compliance costs, spread the refund claims and displace the fraud to other sectors or goods. The dynamic nature of fraud is thus likely to create pressure to further extend the scope of the reverse charge mechanism, which, in the end, would convert the VAT into a retail sales tax.30

In order to increase the efficiency and effectiveness of their controls, the EU Member States are in the process of strengthening their anti-fraud strategies by developing systems capable of detecting risky situations and parties. Since missing traders need to obtain a VAT identification number – unless they hijack another trader’s VAT identification number – preventive “risk analysis” has become an essential element for tackling fraud. To that end, many EU Member States24 use software and electronic databases containing “macro” and “micro” data relating to traders, including the economic sector in which they operate, their financial situation and equity ownership, as well as information regarding their history of payments, declarations and refunds of VAT. Risk analysis, intelligence efforts and cooperation with other national bodies and OLAF are important elements of the battle against fraud on the importation of goods from third countries. The law-enforcing measures taken by the customs authorities are commonly based on inspections in border areas and preventive controls based on computer technology for scanning and tracking goods. To avoid “underbilling” mechanisms and stop foreign channels of fraud, the customs authorities avail of regulatory provisions enabling them to adjust the declared value of imported goods on the basis of general information relating to the market price of the goods and specific information relating to individual transactions (transport costs, insurance, and freight).25 Other measures aimed at combating fraud concern the adoption of ad hoc checks of VAT refund claims, refusal of deduction of input VAT, application of “joint and several liability” provisions for payment of VAT, and denial of application of the zero rate to intra-Community supplies of goods. Such measures were introduced in many EU Member States26 and were generally aimed at specific sectors (computers, mobile telephones and other electronic devices) and transactions for a price below the “open market value” of the goods.27 However, the effect of those measures can be weakened or nullified, if the traders’ VAT claims are offset against their other tax liabilities or if the refund is claimed (or the zero-rated intra-Community supply is made) by traders unwittingly involved in the fraud. In the latter case, the concept of “economic activity”,28 as interpreted by the European Court of Justice (ECJ), and the Community 336

INTERNATIONAL VAT MONITOR SEPTEMBER/OCTOBER 2008

However, the strengthening of the Community legal framework has not been followed by an appropriate response by the Member States. As regards the efficiency of the cooperation between the Member States, the European Court of Auditors31 has judged that the administrative structures and operational procedures are inadequate, the Member States’ replies to requests for

23. A network – used by 24 Member States with the aim of making easier exchanges of information on traders suspected of being involved in carousel fraud – creates a further form of cooperation with respect to those established by the aforementioned Regulation. 24. For example, in the United Kingdom, the Netherlands, Belgium, Spain, France, Italy and Hungary. 25. In Italy, for example, as from 2006, the customs agency is allowed to acquire data and documents relating to each element of cost that forms the declared value of an import. The request for information may apply to all parties potentially involved in the fraud, including importers, exporters, airport service companies, shipping companies, and individuals who handle, deposit and transport the goods. 26. For example, in the United Kingdom, the Netherlands, Germany, France, Spain and Italy. In France, the tax authorities may, with effect from 2007, deny the supplier the right to apply the zero rate to intra-Community supplies, if he knew (or had reasonable grounds to suspect) that the buyer carried out simulated activities. 27. The legal basis of such provisions is Art. 206 of VAT Directive 112/2006, which allows Member States to “(...) provide that a person other than the person liable for payment of VAT is to be held jointly and severally liable for payment of VAT”. 28. The concept of “economic activity” is not dependent on the possible fraudulent nature of another transaction in the chain (ECJ judgment of 12 January 2006 in Optigen Ltd, Fulcrum Electronics Ltd and Bond House Systems Ltd v. Commissioners of Customs and Excise, joined Cases C-354/03, C-355/03 and C-484/03, [2006] ECR I-483) and the right to deduct VAT can be subject to limitations only where the taxable person knew (or had reasonable grounds to suspect) that VAT would go unpaid on the supply to which it is party or earlier in the supply chain (ECJ judgements of 6 July 2006 in Axel Kittel v. Belgian State, joined Cases C-439/04 and C-440/04, [2006] ECR I-6161 and of 11 May 2006 in Commissioners of Customs and Excise v. Federation of Technological Industries and Others, Case C-384/04, [2006] ECR I-4191). 29. It is precluded to make any general distinction between lawful and unlawful transactions for the purposes of levying VAT and to treat similar economic transactions, which are therefore in competition with each other, differently for VAT purposes (ECJ judgements of 22 May 2008 in Ampliscientifica Srl, Amplifin SpA v. Ministero dell’Economia e delle Finanze, Agenzia delle Entrate, Case C-162/07, of 23 October 2003 in Commission of the European Communities v. Federal Republic of Germany, Case C-109/02, [2003] ECR I-12691, of 16 September 2004 in Cimber Air A/S v. Skatteministeriet, Case C-382/02, [2004] ECR I-8379 and of 18 October 2007 in Navicon SA v. Administración del Estadocase, Case C-97/06, [2007] ECR I-8755). 30. M. Keen and S. Smith, “VAT Fraud and Evasion: What Do We Know, and What Can Be Done?”, see note 2. 31. “Special report No. 8/2007 concerning administrative cooperation in the field of value added tax, together with the Commission’s replies”, OJ C20/1 of 25 January 2008.

© IBFD

Articles

information from other Member States are too late in comparison with the prescribed time frames, the information received by the Member States is not always reliable and it is not systematically used. Delays in responding to requests for information and weaknesses in the monitoring system and in the organizational set-up are due to a lack of priority given by local tax offices to those requests. Regarding the VIES, urgent action must be taken to radically shorten the time frame for collecting data and to promptly correct inaccurate information.

ber of traders claiming a VAT refund is equal to those declaring a VAT debt. As regards businesses subject to the “studi di settore”,39 the amount of VAT claimed for 2005 was greater than the total revenue derived from direct taxes, IRAP (a local tax on business activity) and social contributions, which may indicate the existence of widespread fraud. Abuse of the arrangements for offsetting VAT claims against other tax liabilities and claiming refunds has reached an abnormal proportion, not comparable with that in the other Member States.40

As far as customs is concerned, the Court of Auditors32 found that the implementation of the NCTS in several Member States was unsatisfactory. Indeed, the Member States apply the legal provisions differently, have only rudimentary risk management and carry out physical checks on goods in transit merely on an incidental basis. Since the customs authorities of some Member States use more advanced technologies than others, fraudsters may focus on the “weaker ports”. Differences like these – in respect of a system which could reduce fraud by EUR 5 to 10 billion over a five-year period33 – call for an urgent and coordinated action at an EU level, to ensure that changes to the NCTS are prioritized in order to combat import fraud.34

In Belgium,41 the loss of revenue due to organized VAT fraud is estimated at EUR 1.1 billion in 2001. That amount has reportedly decreased to EUR 159 million in 2005, EUR 50 million in 2006 and EUR 44 million in 2007. Since 2001, Belgium has pursued a policy against carousel type of fraud based on a strong and coordinated action against the parties concerned and the creation of a specialized unit (the OCS/TVA) that plays a central role in the battle against VAT fraud: it rapidly forwards to the local tax authorities and legislative authorities all information derived from its investigations.

The differences between the activities of the customs authorities of the individual Member States, together with the difficulties in cross-border prosecution, have prevented the full exploitation of the potential benefits of mutual cooperation. Estimates published over the years show growing EU losses attributable to VAT fraud.35 Useful information emerges from data available at national level, even though it is fragmented and not homogeneous. In Spain, VAT fraud, which amounts to approximately EUR 4 billion per year,36 has been significantly reduced over the last years. Since 2005, this Member State has pursued a “plan for the prevention of tax fraud”, which, as regards VAT, is based on digitization of tax returns (the number of returns filed in electronic format has increased from 23.3% in 2005 to 35.2% in 2007), reallocation of internal resources and enhancement of preventive risk analysis. Refund claims made by traders acting in “risky” sectors have reduced by 21% in 2005 and by 53% in 2006.37 In 2006, this focus on “risk areas” enabled the authorities to detect fraud networks, money laundering and false invoicing for about EUR 3 billion. In 2006, thanks to preventive controls, audits of persons involved in fraud have dropped by more than 40%. During that year, preliminary checks (before registration) have been made on more than 13,000 companies, and registration in the register of intra-Community operators has been denied to more than 1,800 applicants. Registrations in that register were reduced by 30% from 2005 to 2008.38 In Italy, the available data is more limited. From January to September 2007, 227 audits revealed fraud amounting to almost EUR 33 million, which is equivalent to 6.7% of the whole VAT evasion detected in the course of about 8,500 regular and targeted tax audits. Despite these efforts, VAT fraud remains significant in Italy. The num© IBFD

More detailed information on VAT fraud and its impact on revenue is published in the United Kingdom by HM Revenue & Customs (HMRC) and the Office for National Statistics (ONS). According to HMRC,42 attempted fraud was GBP 1.2-2.3 billion in 1999/2000 and remained quite constant until 2002/03 (GBP 1.5 to 2.3 billion), but fell in 2003/04 (GBP 1.1 to GBP 1.7 billion). This reduction was achieved through a range of activities, including undertaking nearly 7,000 preregistration visits to new traders, refusal of more than 900 suspect registrations and cancellation of nearly 500 registrations. In subsequent years, attempted fraud increased again to GBP 1.1-1.9 billion (2004/05) and GBP 3.5-4.7 billion (2005/06), and dropped again in 2006/07 (GBP 2.2-3.2 billion). A significant part of the

32. Court of Auditors, “Special report No. 11/2006 on the Community transit system, together with the Commission’s replies”, see note 17. 33. Resolution of the European Parliament of 11 October 2007. 34. House of Lords, see note 8. 35. These estimates undoubtedly may reflect a growing awareness of the phenomenon by the Community authorities rather than a real increase in the level of evasion. It is also clear that the estimates are the result of the inclusion of new types of fraud previously unknown and evasive behaviour, which are not the focus of this article. 36. Agencia Tributaria, Observatorio del delito fiscal, Primer informe, December 2006. 37. Agencia Tributaria, Balance de actuaciones del Plan de Prevención del Fraude Fiscal, February 2008, www.agenciatributaria.es. 38. Agencia Tributaria, press release, 18 July 2007, www.agenciatributaria.es. 39. The “studi di settore” are a form of inductive investigations based on statistics of taxable persons (essentially SMEs and professionals). 40. V. Visco, Relazione al Parlamento del Ministro dell’Economia e delle Finanze. I risultati della lotta all’evasione, 22 October 2007, www.camera.it., p. 27. 41. Rapport d’activités de la police fédérale (2007), pp. 23-24, www.polfedfedpol.be. 42. HM Revenue & Customs, “Measuring and tackling indirect tax losses, years 2004-2007” (2005). HMRC currently produces a bottom-up estimate based on operational evidence that provides a range for both attempted (rather than actual) missing trader intra-Community fraud (MTIC) and effect on VAT revenue (the methodology is not revealed in order to avoid a detrimental effect on compliance activity).

INTERNATIONAL VAT MONITOR SEPTEMBER/OCTOBER 2008

337

Articles Figure 2

attempted fraud could be stopped and, therefore, the impact of VAT fraud on VAT revenue is substantially lower (GBP 2 to 3 billion in 2005/06 and GBP 1 to 2 billion in 2006/07).

suspect VAT repayment claims)49 and the government’s new strategy based on the legislation that enables the tax authorities to hold traders jointly and severally liable for payment of unpaid VAT debts.

The sudden and large increase in attempted fraud during the years 2005/06 must partly be attributed to the judgment of the ECJ43 in which the ECJ ruled against the approach taken by the United Kingdom to deny suspect VAT repayment claims made by taxable persons involved in carousel fraud.44

It should, however, be noted that the fall in VAT fraud registered by the ONS only concerns the sectors of computer chips and mobile telephones, i.e. sectors where, from June 2007, the reverse charge mechanism has been applicable to all domestic B2B supplies exceeding GBP 5,000. As the introduction of the reverse charge mechanism had been announced several months in advance, the measure may have produced an anticipatory effect in the period prior to June 2007. Moreover, trade flows have been considerably distorted in the period 2005/06 due to the escalation of MTIC fraud and in particular by a huge and exceptional carousel fraud estimated at several billions of euro and involving a large number of Member States.50 Hence, considerable uncertainty remains about whether the actual fraud is fully reflected

The monthly data published by the Office for National Statistics45 shows a sharp growth in trading activities associated with VAT fraud (MTIC fraud adjustment) from GBP 2.6 billion in 2004 (GBP 0.65 billion per quarter) to GBP 11.2 billion in 2005 (GBP 1 billion in the first, GBP 2.3 billion in the second, GBP 3.5 billion in the third and GBP 4.4 billion in the last quarter) and GBP 20.7 billion in the first half of 200646 (GBP 10 billion in the first and GBP 10.7 billion in the second quarter). If maintained throughout the year, the figure for the first half of 2006 would have implied a revenue loss of some GBP 4.5 billion, about twice the upper limit of HMRC’s estimate for 2004/05.47 The ONS subsequently registered a fall of fraud-related trading activity, to about GBP 100 million per month (see Figure 2, on which quarterly flows are charted). The sharp increase in the MTIC fraud adjustment in 2005/06 has only given rise to a slight dip in VAT revenues.48 It may possibly be explained by the enhancement of HMRC’s controls (including rigorous checks on suspect applications for VAT registration and the use of dedicated teams to carry out in-depth verification of 338

INTERNATIONAL VAT MONITOR SEPTEMBER/OCTOBER 2008

43. Joined Cases C-354/03, C-355/03 and C-484/03, see note 28. 44. Supplementary letter by Rt Hon Dawn Primarolo MP, Paymaster General, HM Treasury, in House of Lords (see note 8), pp. 61 and 62. 45. Import figures for trade in goods published by the Office for National Statistics include, since May 2003, adjustments to allow for the impact of VAT MTIC fraud. These estimations on MTIC-related activity use overseas trade data supplied by HMRC. 46. Office for National Statistics, UK trade, April 2008, ONS. 47. M. Keen and S. Smith, “VAT Fraud and Evasion: What Do We Know, and What Can Be Done?”, see note 2. 48. UK Statistics Authority, Time series data – UK Economic accounts, Central Government, time series data, www.statistics.gov.uk. 49. HM Revenue & Customs, “Protecting tax revenues”, March 2008. 50. Eurojust, press releases, The Hague, 13 March 2007; HM Revenue & Customs, “Measuring indirect tax losses 2007”, October 2007.

© IBFD

Articles

in the official data.51 The reverse charge mechanism prevents fraud in sectors where fraud is substantial and occurs frequently, but fraud can easily migrate to other industrial sectors, and that migration gives rise to a subsequent increase in the level of fraud.52 Solid and strongly structured organizations “manage every time to renew their circuits of fraud, despite the legal and fiscal proceedings taken against them”.53 4. Towards a New Integrated Strategy In summary, the fight against VAT fraud is far from being won. Extremely flexible criminal schemes only encounter a fragmented anti-fraud strategy of the authorities both at Community and national levels. However, the current Community efforts to combat fraud are focused on developing a more coordinated strategy, which continues to rely on conventional measures aimed at strengthening the VAT system without altering its principles.54 At a later stage, these efforts may be combined with more ambitious measures which have the effect of modifying the VAT system itself. Improving the exchange of information between the Member States entails the adoption of several measures, such as extension of the VIES system to services in respect of which VAT is due under the reverse charge mechanism, reduction to one month of both the frequency of recapitulative statements and the deadline for the exchange of information between Member States,55 automated access of Member States to specific data relating to taxable persons and their activities located in another Member State, creation of a European network in which officials from the national tax authorities participate and, finally, exchange of information with nonEU countries. Moreover, a successful strategy needs a multidisciplinary approach, which includes the enhancement of the intelligence units, for easier access to information. Following the main ECJ case law on this topic, an important aim is to provide a legal base for making traders established in one Member State jointly and severally liable for tax losses occurring in another Member State as a result of non-compliance of the trader with his reporting obligations. The mechanism would enable an easier prosecution of traders aware of the fraud, who do not file VAT returns in order to dissimulate fraud. A more complex but essential goal is the setting up of a “risk analysis” system at EU level, which would have a synergetic effect by combining the experience of the tax authorities of the individual Member States. IntraCommunity fraud can not effectively be fought with a national approach, and businesses and tax authorities have a common interest in this matter. Unilateral actions taken by some tax authorities put a disproportionate burden on businesses without being successful in tackling VAT fraud. Effective consultation between the tax authorities for the purposes of exchanging best practices, and adoption of a common minimum standard for supervision of registration and deregistration of taxable persons are therefore essential. With regard to customs, a © IBFD

proper balance between trade facilitation and effective protection of the Community’s financial interests calls for appropriate risk management and risk-based checking of transits.56 Achieving a European risk analysis requires a wide integration of available databases and the development of more detailed and uniform risk parameters. Identifying the flows of money used to start (or derived from) VAT carousels is crucial to detect the associated money laundering. A key element is an increase in the global awareness of the fraud mechanisms and their links with other crimes: financial institutions may receive the proceeds of VAT carousel fraud from a government source in the form of a VAT repayment and it may not arouse any suspicion; only a good knowledge of the main fraud schemes and the nature of the businesses carried out by the addressee may reveal an inconsistency between the size of the financial transactions and the addressee’s commercial activity.57 Once more, it appears necessary to strive for a closer interaction between the regulated sector, the Financial Intelligence Units (FIUs) and the competent authorities, together with the creation of international risk profiles. However, an integrated EU strategy capable of producing sizeable results in the fight against VAT fraud may call for far-reaching changes in the VAT system, such as taxation of intra-Community supplies of goods or introduction of a generalized reverse charge mechanism. It cannot be denied that pathologic phenomena, such as VAT carousel fraud, have been facilitated by the coexistence of different VAT regimes in the European Union (the regime applicable to intra-Community supplies of goods differs from that applicable to domestic transactions). Adoption of the “origin principle” – as advocated by the EU Commission since 1987 as the only way to ensure the creation of a real internal market – would provide a structural answer but that solution is still blocked, unless some sensitive issues are resolved:

51. “Survey indicators have continued to point to strong goods exports growth (...). That contrasts with official data, which suggested a sharp slowing in the first half of the year”. Bank of England, “Inflation Report”, November 2007, www.bankofengland.co.uk/publications/inflationreport. 52. House of Lords, see note 8. 53. Police fédérale, see note 41. 54. European Commission, “Communication from the Commission to the Council concerning some key elements contributing to the establishment of the VAT anti-fraud strategy within the EU”, COM(2007) 758, 23 November 2007; European Commission, “Oral report by the Commission on conventional measures”, ECOFIN 14 May 2008. 55. The change of the frequency concerns a limited number of businesses (only 4% of businesses liable for VAT submit recapitulative statements and 9% engage in intra-Community acquisitions of goods) and the additional cost will be very small. See European Commission, “Proposal for a Council directive amending Directive 2006/112/EC on the common system of value added tax to combat tax evasion connected with intra-Community transactions – Proposal, for a Council regulation amending Regulation (EC) No. 1798/2003 to combat tax evasion connected with intra-Community transactions”, COM(2008) 147, 17 March 2008. 56. Court of Auditors, “Special report No. 11/2006 on the Community transit system, together with the Commission’s replies”, see note 17. 57. Financial Action Task Force, see note 5.

INTERNATIONAL VAT MONITOR SEPTEMBER/OCTOBER 2008

339

Articles Figure 3





introduction of a “clearing house system” to adjust the revenue imbalance (in the absence of such a system, Member States with a trade surplus would obtain a significant advantage at the expense of Member States with a trade deficit); the question that remains unsolved is whether the system should be based on business-level (microeconomic) recapitulative statements or on a macroeconomic bilateral clearing system (based on official trade statistics); more uniformity between national VAT regimes, in order to avoid competitive distortions in favour of products from Member States with lower VAT rates. The goal is still far away: in January 2008, the average standard rate of VAT was equivalent to 19.5%, ranging from 15% in Cyprus and Luxembourg to 25% in Denmark and Sweden. The average reduced rate was 8.5%, ranging from 5% to 17%. Nevertheless, the sharp reduction of the coefficient of variation between the standard rates (from 0.20 to 0.12 in the period from 1993 to 2008) reveals a remarkable convergence (see Figure 3).58

The unanimity required for all common tax measures creates huge inertia in dealing with coordination problems of this kind, in particular between Member States who are not willing to give up their tax sovereignty.59 The Commission has recently proposed a “flat-rate origin system”,60 similar to a “cross-border VIVAT”,61 that does not require a far-reaching harmonization and, yet, avoids distortions of competition. Under the proposal, the zero rate for intra-Community supplies of goods would be replaced by taxation at the rate of 15% and the domestic VAT rates would continue to apply by assessment of the recipient of the goods for the difference.62 That proposal is accompanied by a proposed microeconomic, bilateral “clearing system”, based on recapitulative 340

INTERNATIONAL VAT MONITOR SEPTEMBER/OCTOBER 2008

statements; this would be quite simple as it is associated with a single VAT rate. Depending on their relative trade balance position, the Member States would either have to pay to, or receive a sum of money from, the other Member States.63 Even though the new system would create a cash flow disadvantage for traders, which may cause difficulties, in particular for SMEs,64 the impact on the costs for businesses would be very small. A major issue for the Member States is that their VAT revenues become dependent on transfers made by the other Member States, which gives rise to the question of whether or not the Member States are prepared to accept that consequence. More importantly, taxation of intra-Community supplies would not provide a final solution to all forms of VAT fraud. Putting an end to fraud relating to zerorated, cross-border transactions may give rise to further 58. Elaborations on European Commission data (see European Commission, “VAT Rates Applied in the Member States of the European Community”, 2008). The average rate has been calculated over 27 countries throughout the whole period 1993- 2008. 59. M. Keen and S. Smith, “VAT Fraud and Evasion: What Do We Know, and What Can Be Done?”, see note 2. 60. European Commission, “Communication from the Commission to the Council and the European Parliament on measures to change the VAT system to fight fraud”, COM(2008) 109, 22 February 2008. 61. M. Keen and S. Smith, 1996, “The future of Value-Added Tax in the European Union”, Economic Policy 23, pp. 373-411 and 419-20. 62. If the Member State of arrival of the goods applies a higher rate than 15%, the additional VAT will accrue to it; on the contrary, if the supply is subject to reduced VAT rates, the Member State of the purchases would make a refund to the taxable person making the acquisition. 63. According to the overall balance of trade statistics, 16 Member States would be “net receivers” whereas the other 11 would be “net payers”. Under a VAT rate of 15%, the “net receivers” Member States would be expecting to receive about EUR 30 billion in tax revenue from the other Member States. European Commission, see note 60. 64. Traders effecting intra-Community acquisitions would have to prefinance the VAT, whereas they currently do not effectively pay VAT.

© IBFD

Articles

development of other, existing forms of fraud, and the overall result is uncertain at this stage: fraud could still be committed by issuing false invoices relating to intraCommunity transactions and rate differentials might be exploited for fraudulent purposes.65 The main objection to the adoption of the origin principle as an anti-fraud strategy is that it would not be effective in combating fraudulent schemes involving traders in third countries. Changing the VAT system without strengthening the system of customs control and, in general, without full consciousness of the likely consequences in terms of displacement of fraud would therefore be illogical. Any overall strategy to combat fraud must be designed in coherence with the context in which it occurs, taking into account any possible alternative or complementary mechanisms prevailing at EU and national level. As regards the scope of the clearing system, carousel fraud makes it difficult to provide accurate data and reliable estimates of the actual trade flows within the Community, which means that there could be a high risk of disputes between the Member States as regards the amount of VAT to be cleared. The Member States would need to clear their VAT data bases of registered businesses, their auditing procedures and the accuracy of the cross-border exchange of information with other Member States to achieve a more efficient VAT system. Thus, reduction of VAT fraud to an acceptable level appears to be a crucial prerequisite for accomplishing an internal market with taxation under the origin principle.66 The problems associated with taxation of intraCommunity transactions have also led to a proposed solution that is diametrically opposite to flat-rate taxation of intra-Community supplies of goods: adoption of a generalized reverse charge mechanism (shifting the tax liability from the supplier to the recipient) in respect of all B2B domestic and cross-border supplies within the European Union. That solution would drastically reduce all fraud at the intermediate stages of distribution. Moreover, a blanket reversal of the tax liability would minimize the disadvantages of the current sectoral reverse charge (i.e. an increase of refund claims and the displacing of fraud in other sectors). The topic poses additional problems that need to be examined further. Notably, a generalized reverse charge mechanism is not an answer to “black sales” and may generate other types of “consumption fraud”: false statements about the customer’s VAT status and the hijacking of VAT identification numbers at the retail level (in order to avoid the supplier having to charge VAT and to allow goods to enter into the black market free of VAT) will also have a negative effect on the Member States’ revenues. The potential losses could be higher if compared to the current system because VAT would no longer be collected in a fractionated manner within the supply chain, and any fraud at the stage of consumption would affect the taxation of the entire added value.

© IBFD

Currently, a large proportion (80%) of the VAT collected by the Member States is remitted by less than 10% of the taxable persons; this means that the tax authorities of the Member States need to invest very little control effort in order to safeguard the bulk of their VAT revenue.67 A generalized reverse charge system would entail a lower risk of fraud in sectors dominated by large businesses that guarantee the levy (for example, large retailers). By contrast, such a system may have negative consequences in sectors where goods and services are sold to final consumers by small businesses and liberal professionals; in countries characterized by a fragmented distribution structure, the reverse charge mechanism considerably weakens the control of the authorities because the tax debt, which is due at the end of the distribution chain, would have to be collected from a larger number of retailers. As compared to the US “retail sales tax”, the incentive to evade the tax would be higher in the European Union because the tax rates are higher. A second-best solution could be the introduction of a “hybrid reverse charge mechanism”, limited to supplies exceeding a certain threshold value, which would combine the advantages of a “cashless VAT system” (under which fraudsters would not be able to charge VAT in respect of their supplies) with the benefits of the built-in safety mechanisms of the “classical” VAT system (that would still be applicable to the bulk of the transactions).68 The EU authorities have examined the implications of such a change69 and concluded that, since businesses would be required to fulfil extra reporting obligations,70 the hybrid reverse charge mechanism would represent significant one-off as well as recurrent costs for traders, which are unnecessary for traders operating in sectors where fraud is not prevalent (the one-off costs for SMEs would vary between EUR 12.750 and 20.000, whilst the annual costs would range between EUR 6.000 and

65. Intra-Community buyers which are resident in a high-rate Member State (e.g. 25%) could omit the payment of the differential rate (10%) and apply lower prices to consumption or start a carousel fraud. See: European Commission, “Commission staff working paper on measures to change the VAT system to fight fraud”, SEC(2008) 249, page 7. 66. A. Gebauer, C. Woon Nam and R. Parsche, VAT evasion and its consequence for macroeconomic clearing in the EU, Institute for Economic Research, 2005. 67. European Commission, “Communication from the Commission to the Council in accordance with Art. 27(3) of Directive 77/388/EEC”, COM (2006) 404/F, p. 4. 68. M. Tumpel, “A Hybrid VAT System in the European Union”, Tax Notes International, Vol. 47 (2007), No. 2, pp. 165-183. 69. European Commission, “Possible introduction of an optimal reverse charge mechanism for VAT. Impact on businesses”, Consultation paper, 13 August 2007. 70. In order to enable the tax authorities to monitor the “VAT-free” domestic and cross-border transactions in goods, the suppliers must provide a periodic list of their customers and the value of the supplies made to them and check the validity of their customers’ VAT identification numbers and the customers must provide a list containing the VAT identification numbers of their suppliers and the value of the goods received from them.

INTERNATIONAL VAT MONITOR SEPTEMBER/OCTOBER 2008

341

Articles

9.30071), in particular for traders with wide distributive chains.72 A hybrid reverse charge mechanism would also give rise to other issues. Firstly, it would lead to a substantial increase of VAT refund claims by taxable persons, who pay VAT to their suppliers (on their inputs not exceeding the threshold value), where their output transactions are subject to the reverse charge mechanism. Secondly, treating identical transactions differently for VAT purposes reduces the effectiveness of anti-fraud measures, increases the complexity of the management of the VAT system and requires a tough control system in order to protect Member States from adverse effects.

For these reasons, the EU Commission takes the view that “a general reverse charge system should either be introduced on a mandatory basis throughout the EU or be disregarded as a concept”73. In any case, the coexistence within the European Union of different VAT systems would be inevitable. The fight against fraud will continue to be delegated to administrative cooperation and national control mechanisms. Once more, a unified strategy based on the interactions between available instruments is a prerequisite.

5. Conclusions The alarming development of VAT fraud demonstrates the inadequacy of national fraudcombating measures and the deficiency of mutual assistance between the Member States. A drastic change of the VAT system could provide a more robust defence against fraud as compared to administrative solutions, but this change is difficult to make and may damage the principles on which the VAT system is based. Moreover, in the absence of appropriate complementary EU and national measures, the effects of such far-reaching measures remain uncertain. Further enhancement of preventive controls aimed at discovering fraud more rapidly and tackling it more effectively is required. A system for risk analysis must

be developed at EU level, using standardized technologies, shared databases and refined risk parameters specific to the transit of goods from third countries. The need to maximize the effectiveness of the strategy to combat VAT fraud cannot be separated from a broad view on the functioning of the VAT system as a whole. Available mechanisms interact, require a high degree of adaptation to the economic and legal context and must be a constant of the Member States’ regulatory framework. This means that there is not a “single road” to a more efficient VAT system and that it is inappropriate for Member States to simply copy and transpose the measures taken by other Member States into their national legislation.

71. Study for the Vienna Chamber of Commerce by the Austrian Institute for SME Research, mentioned in document COM(2008) 109, see note 60. 72. PricewaterhouseCoopers, “Study in respect of introducing an optimal reverse charge mechanism in the EU VAT Directive”, Final report to the EC, PWC, 20 June 2007. 73. European Commission, COM(2008) 109, see note 60.

342

INTERNATIONAL VAT MONITOR SEPTEMBER/OCTOBER 2008

© IBFD